Indian Journal of Experimental Biology Vol. 37, February 1999, pp. 11 7-123

Review Article

Experimental methods for evaluation of psychotropic agents in rodents: II-

S K Bhattacharya, K S Satyan & M Ramanathan Neurophannacology Laboratory, Department of Phannacology, Institute of Medical Sciences, Banaras Hindu Un iversity, Varanasi 221 005, India

Rod ent model s of clinical 'are extensively used for the evaluation of putative antidepressants. In the present review, the available experimental methods which can be utilized by most laboratori es involved in preclinical screening of antidepressants, have been discussed. The methods have been categorized on the basis of induction of the depressive state or on th e assumption that monoamine deficiency leads to depression. These methods have been critically validated in terms of efficacy of standard antidepressants in these tests and, in some cases, by th e neurochemical basis of depression, namely, the deficient theory of clinical depression.

Depression is a chronic illness that affects people of all elled in animals. Some criteria have been suggested for the ages. The lifetime risk of major depression disorders in validation of animal models of psychiatric disorders, community samples varies from 10-25% for females and 5- namely predictive, face and construct validity. These have 12% for males I. Although there are many effective antide­ been discussed at length in a recent review on evaluation of pressants available today, the current armentarium of ther­ agents'. In brief, the animal mode! used should apy is often inadequate, with unsatisfactory results in about be able to predict clinical use, have some behavioural and l one third of all subjects treated . This provides impetus in neurochemical similarity with clinical depression, and the search for newer and more effective antidepressants. As clinically effective antidepressants should be effective in our understanding of the biochemical basis of endogenous the animal model as well. Thus, these models cannot only depression proceeds beyond the confmes of the classical predict activity, but also provide an insight monoaminergic deficiency hypothesis, to involve other into the psychobiology of depression. A variety of animal neuro-transmitter systems, including , choli­ models can be used to detect antidepressant activity . It is nergic and peptidergic neuronal activity I, the search for always imperative that a battery of tests be used to evaluate newer antidepressants continues to expand and diversify. and confirm preclinical antidepressant action. The large number of new antidepressants introduced in recent years into clinical practice have, unfortunately, Methods failed to surpass the plateau of efficacy exhibited by the Animals-Adult rats and mice are used. Several investi­ classical antidepressant agents, the monoamine oxidase gators have used only male rodents, although there does (MAO) inhibitors and antidepressants (TeAs). not appear to be any significant degrees of sex variation in 4 Ideally, a new antidepressant should not only have greater the induction of depression models . However, it is always efficacy and be cost-effective, it has to have a rapid onset advisable to ensure that the animals belong to a specific 4 of action. The expected delay in clinical efficacy of most of species and are inbred, to avoid variability of the results • the antidepressants available today is between 10 days to 3 Drugs, route of administrative and duration of treat­ weeks I. The search for such an ideal antidepressant as a ment-It is always advisable to administer the test drug more effective medication, with better tolerability, rapid orally, in graded doses, since this wili be likely route of onset of action and ease of dosing, requires valid preclini­ clinical use. However, in case of a new drug, with un­ cal tests which can predict clinical antidepressant activity. kn9Wll pharmacokinetic profile, the intrapel itoneal route The plant kingdom may be a part of this search for the can be used for initial experimentation. A vehicle treated 'Holy Grail' of antidepressant psychopharmacology, as control group is important for the sake of statistical evalua­ evidenced by the emergence of Hypericum per/oratum (St. tion. Single acute drug administrati on can be adopted. John's wort) as an effective clinical antidepressanf. However, in case of plant extracts, subchronic (3-5 days) drug administration may be required. . Like any other field in psycho-pharmacology, preclini­ cal research on anti-depressants is confounded by the vir­ Standard antidepressants- The standard drug used i tual impossibility of developing investigative animal mod­ usuall y ( 10 mg/kg, ip pretreatment time 30-45 els of clinical depression. The major problem faced by all min). However, a MAO lIlhibitor like (1 0 mg/kg, anirnal modeis of depression is that core features of depres­ ip pretreat:rrient time 2 hr) may be used if this is the likely sion - the depressed mood, feeling sad, neglected and mel­ mode of action of the test drug. It is important to note mat, ancholic, as well as the suicidal tendency, cannot be mod- while classical tricyclic antidepressants like imipramiIli: can 118 [NOlAN J EXP BIOL , FEBRUARY 1999 be effective both on pre- and post-treatment, the MAO in­ actions of , including hyperthermi<1, aug­ hibitors are effective only on pretreatmen~. mented locomotor activity, stereotypy and lethality in ag­ Animal models of depression-Basically two types of gregated rodents 9 rodent behaviour models are used. The first type uses drugs a. Amphetamine-induced stereotypy----Amphetamine (5- to induce depression or depends upon drug induced be­ 10 mg/kg, ip) induces a stereotyped behaviour which is 51 ha viours, and the other type utilizes behavioural models best noted 30 and 60 min after adminisu'ation in rats . 0. which have some resemblance to endogenous depression. The rat is placed in a spacious cage in a dimly-lit quiet room. The latency of onset, intensity and duration of I. Pharmacological models stereotypy, are assessed and scored- (A) Reserpine (or tetrabenazine) syndrome-Reserpine 1. Discontinuous sniffing, constant exploratory activity depletes central and peripheral monoamines, whereas tetra­ 2. Continuous sniffing, periodic exploratory activity, benazine has a selective central action. These drugs induce small head movements a syndrome (ptosis, hypothermia, catalepsy and decreased 3. Continuous sniffing, small body and head move­ locomotor acti vity), the reversal of which is used as a reli­ ments, discontinuous gnawing, biting and licking the cage able initial method to detect antidepressant activity. Reser­ wall, brief spurts of locomotor activity pine (2.5-5.0 mg/kg, sc) or tetrabenazine (10 mg/kg, ip) are 4. Continuous gnawing, biting and licking of cage wall, used and the pham1acological effects are assessed 2 hr no ambulation except for occasional backward move­ lO later". TCAs are effective in attenuating the syndrome both ments . on pre- and post-treatment, whereas MAO inhibitors are b. Amphetamine-induced' hyperthermia-Temperature effective only on pretreatment5. In addition, the proconvul­ changes are recorded at 30 min illtervals for 2-4 hr after sant, blockade of conditioned avoidance response" in ro­ administration of amphetamine (2.5-5.0 mg/kg, ip) by a dents, and emetic (pecking) response in pigeons, have also telethermometer, as mentioned earlier, recording rectal 5 1O been used as test parameters. temperature . • a. Reserpine-induced ptosis-Ptosis or palpebral closure c. Amphetamine-induced increase in locomotor activ­ is graded from 0 to 4, 0 being complete closure and 4 indi­ ity---Locomotor activity is recorded in an automated activ­ 6 cating that the eyes are widely open . ity cage, as mentioned earlier, at 10 min intervals for 30 b. Reserpine-induced decrease in locomotor acti­ min, 2 hr after amphetamine (1-2 mg/kg, ip)5. vity-Reduction in locomotor activity is best tested in an d. Amphetamine- induced toxicity in grouped ro­ au to mated acti vity cage recording photobeam breaks as dents-Mice are usually used5 and groups of 10 mice are 3 counts. Rats are placed individually and 10 min counts are kept crowded in a small wire mesh cage (16 cm ). 30 min recorded for 30 min, to obviate the initial locomotor spurt later amphetamine (10 mg/kg, ip) is administered to the in ambulatory behaviour5 animals and lethality is recorded 4 hr later. Results are ex­ c. Reserpine-induced hypothermia- Rectal temperature, pressed as percentage increase in amphetamine lethality as considered as the core body temperature, can be recorded compared to vehicle-treated controls. This dose of am­ 5 8 using a multichannel telethermometer . The animal is phetamine induces 20-30% mortality in mice . The rela­ placed in a loosely fitting perspex chamber and a themlis­ tively newer antidepressants, and trazadone are q to r probe i~ inserted 4 em deep into the rectum (in rats), ineffective in this test . and kept in situ for the duration of the experiment. Tem­ (C) antagonism-Apomorphine in higher perature responses are noted before and I, 2 and 4 hr after doses (16 mg/kg, ip) induces hypothermia which is not 5 reserpine administration . antagonized by receptor blocking neuroleptics d. Reserpine-induced catalepsy----A variety of methods which can, however, attenuate apomorphine-induced can be used to assess catalepsy. The Pertwee's ring tese has stereotypy and climbing behaviour. A wide range of anti­ been shown to be sensitive in distinguishing between MAO depressants can, on the contrary, reverse the hypothermia. 5 A and MAO B inhibitors . The rat is placed on an iron ring This method is simple and useful for rapid detection of (diameter 12 cm) fixed to a steel stand at a height of 15 cm. antidepressant acti vity 1 I . The time during which the rat remains motionless, with the (D) Potentiation of -induced convul­ complete cessatio n of snout and whisker movements, out of sions-Tryptamine (60 mg/kg, ip) produces bilateral clonic a tota l observation period of 5 min, is used to calculate per convulsive movements of fore paws in mice, characterized cent inmlobility. A smaller ring (6 cm) can be used for by pronounced up and down movements of the paws which mice. push the animals backwarcis,resulting in retro-pul ion. Te As and tvlAO inhibitors potentiate a dose of tryptamine The method is simple, rapid and reliable, and can detect 8 ( 15 mgikg, ip) which produces minimal convulsions . al l classes of antidepressants. Rats provide more cogent data than mice8 However, false negatives (mianserin) and (E) potentiation-Yohimbine (25 mg/kg, false positives (, antihistaminics) are on rec­ sc) an antagonist at alpha-2 receptors, produces Q ord • minimal lethaliiy in rodents. However, TCAs, MAO in­ (8) Amphetamine potentiation-Most antidepressants, hibitors and most of th e newer antidepressants, potenti ate including TCAs and MAO inhibitors potentiate the central the le thality of this dose of yohimbine in rats and mice9 BHA ITACHARY A et al.: EV ALVA TlON OF PSYCHOTROPIC AGENTS 119

This test has been used as a simple and rapid test for inducing motor deficit. A major precaution is to remove the screening diverse groups of antidepressants. mouse ca,rcas~ promptly to prevent it being eaten up by the (F) Potentiation of 5-hydroxytryptophan (5-HTP) re­ killer rat. Rats exhibiting muricidal behaviour can be re­ sponses--5-HTP, the 5-HT precursor, produces typical used but they have to be caged individually in isolationu. behavioural responses in rodents. The effect in rats, desig­ Non-muricidal rats can be rendered muricidal by pretreat­ nated 'wet dog shakes', comprising of intermittent body ment with pilocarpine (2.5 - 5.0 mg/kg, iPt movements including a shaking of the head, whereas, in mice, the effect is predominantly rapid and intermittent U. Behavioural models head-twitches 12.13. In mice, a dose of 5HTP (50 mg/kg, ip) These models are based on manipulations of social rein­ which produces minimal head-twitch response is used, and forcement or environmental unpredictability. The induced the potentiation induced by drugs is noted by counting the behavioural responses of the animal is postulated to repre­ twitches at three two minute intervals (19-21 , 23-25 and sent clinical depression 16. 27-29 min) after 5-HTP administration. The fmal data is (A) 'Behavioural despair' test 16.17_This model is presented as the mean of the head-twitches during the test based on the premise that, when rats or mice are forced to periods8. This model can predict anti-depressants influenc­ swim in an apparatus from which there is no escape, they ing 5-HT activity, namely selective 5-HT reuptake inhibi­ will, after initial frenzied attempts to escape, adapt a char­ tors, like , and 5-HT selective TCAs, like chlori­ acteristic immobile posture and make no further attempts to mipramine. escape. The movements made by the rodent are those nec­ (G) Post-swim grooming respons~roups of mice are essary to keep its head above the water level. It is postu­ placed in a water bath containing water (32°C, depth 10 lated l6.17 that the immobility exhibited by the animal re­ cm) for 3 min. Thereafter, the animals a~e removed and flects a state of 'despair', resigning itself to the experi­ observed for grooming behaviour, every min, during a 10 mental situation. In a standard protocol, the rat is placed in sec period, for 30 min. A score of one is given if the mouse a plexiglass chamber (45 x 40 x 30 cm) with water level of was grooming during the 10 sec observation period or 25 cm (25±2°C), so that it cannot touch the bottom with its scored as 0 if not grooming. Thus each mouse could exhibit hind paws or tail, and climb over the edge of the apparatus. a maximum score of 30. This test, being a dopamine­ The animal is allowed to swim for 10 min, during which it mediated response, can be used to detect MAO B inhibitor .makes several attempts to escape and swims vigorously. activity8. 14 . Thereafter, the total period of complete cessation of swim­ (H) Potentiation of I-DOPA induced response--Mice ming with the head floating just above the water level, is are treated with I-DOPA (25 mg/kg, ip) and benserazide noted during the next 5 min period. The protocol is same in hydrochloride (6.5 mg/kg, ip). The change in behaviour is mice except that the vessel dimensions (35 x 15 x 10 cm) scored every 10 min for 30 min8. A scoring system, ranging and water level (15 cm) are different. The classical TCAs from I to 4, is used, based on the following criteria: reduce immobility time and a significant correlation exists 1=piloerection , minima! increase in ambulation, between the experimental and clinical potencies of these 2=piloerection, marked ambulation, salivation, irritability, drugs. However, antidepressants acting selectively on the 3=piloerection, profuse salivation, jumping, vocalization, 5-HT system are generally inactive in this test and false increased ambulation, 4=piloerection, profuse salivation, positive are induced by opiates and anti-histarninics. stereotypy (compulsive gnawing and biting the cage wall), (B) 'Learned helplessness' test--This model is based reduced ambulation, aggressive behaviour and automutila­ on the assumption that, exposure to uncontrollable stress tion (biting of tail and fore paws). The dose of I-DOPA associated with repeated experiences of failure to escape used produces minimal behavioural changes, which are from the stress, produces a 'helpless' situation, which re­ potentiated by all classes of antidepressants except those sults in performance deficits in subsequent learning I8 19 acting selectively through increased 5-HT activitl. tasks . . A typical experiment involves two parts: a. lnes~pable shock treatment--Rats are subjected to (/) Muricidal behaviour in rats--Muricidal, or compul­ footshocks in a two compartment jumping box with the si.ve mouse killing behaviour, was noted in female rats of escape route to the adjoining unelectrified 'safe' chamber the Holtzman strain which instinctively attacked and killed closed. A constant current shocker is used to deliver 60 mice irrespective of their satiety status lS . However, a small scrambled shocks (15 sec duration, 0.8 rnA every min) percentage of other rat species can also exhibit muricidal through the steel mesh grid floor. Control animals are behaviour (20% in Charles Foster and Wistar strains). Rats placed in the chamber for 1 hr without experiencing are prescreened for muricidal behaviour, 48 hr before the shocks. This exercise is repeated 48 hr later on day 3. test and the behaviour is confirmed 24 hr later. The per­ centage of rats exhibiting muricidar behaviour within 30 b. Conditioned avoidance training-On day 3, after the sec of introduction of a mouse into the rat cage is noted. second inescapable shock treatment, the rats are subjected Antidepressants attenuate muricidal . behaviour at doses to avoidance training where a rat is placed in the electrified below that inducing motor incoordination (as tested by the chamber and allowed to acclimatize for 5 min before being rota-rod method). Other psychotropic agents, including subjected to 30 avoidance trials, with an inter-trial interval neuroleptics, block muricidal behaviour only at dose levels of 30 sec. During the first 3 sec of each trial, a buzzer 120 INDIAN J, Exp· BIOL,: FEBRUARY 1999 stimulus or a light signal (conditioned stimulus, eS) iii pre­ domly to a variety of stressors during a 3 week period. The sented, followed by footshock (0.8 rnA for 3 sec dur,ation, stressors include mild electric shock, immersion in cold H unconditioned stimulus, UeS). The avoidance response is water and rev<:rsal of light/dark cycle. Another method , characterized by escape to the adjoining unelectrified utilizes a chronic mild stress situation extending over 3-4 chamber during es, and is designated 'escape response'. week period involving exposure of the rat to overcrowding, Failure to exhibit escape response during es is assess:ed as wet saw dust on the cage floor, tilting of the cage, flashes 'escape failure', which is said to represent depressive be­ of light or loud sound, and intermittent removal of food haviour!9. Antidepressants reduce or even eliminate escape and water. However, a much simpler form of chronic un­ 2S failures. This model has excellent predictive valdity and is predictable stress , which consistently induces depression extensively used to screen antidepre-ssants, investigate and other behavioural changes, including attenuation of their mode of action and to evaluate the neurobiology of male sexual activity and cognitive dysfunction, is daily depressive illness! 8. Rats subjected to inescapable shock exposure of rats to I hr of foot or tail shock (2 rnA,duration also exhibit decreased ambulation and aggression, and loss 3-5 sec) for a period of 2-3 weeks. The intervals between of appetite with weight loss, which can be utilized as addi­ the shocks is randomly programmed (unpredictability) tional investigative parameters! 8. between 10 and 100 sec. Control animals a.re placed in the (C) Isolation-induced hyperactivity-Rats are housed test chamber but receive no shocks. All these methods use singly in a cage (38 x 26 x 20 em) in comparison to the the increase in plasma corticosterone as the stress indicator. normal housing of 4-5 'rats/cage. It is ensured that the so­ The methods are sensitive to antidepressants of all classes cially-deprived rats are kept in isolation without visual or and MAO inhibitors. auditory contract with their normally housed counterparts, An alternative chronic stress model of depression,26 for a period of 10- 14 days. Thereafter, the isolated rats are shown to have predictive validity and reliability, but not subjected to behavioural testing. Apart from locomotor construct validity, involves exposure of rats or mice, se­ activity (reduced spontaneous locomotor and exploratory quentially over a period of weeks, to a variety of mild behaviours) and' open-field behaviour (reduced ambula­ stressors, and the measure most commonly used to evaluate tion, rears and immobility), a stereotyping rating, taken for consequent depression is the decrease in consumption of a. 20 10 min, on a 0-6 point scale, has been used . The ratings palatable sweet solution. The generalised decreased in re­ indicate O=inacti vity or sleep, I =active but decreased re­ sponsiveness to rewards is comparabie to anhedonia, the sponse to external stimuli, 2=increased ambulation with core symptom of the melancholic subtype of major depres­ bursts of stereotyped sniffing or rearings, 3=marked sive disorder. In a typical experiment, the rodent is exposed stereotyped behaviour maintained over a wide area of the sequentially to a variety of mild stressors, including over­ cage, 4=marked stereotyped behaviour, mainly sniffing aqd night illumination, periods of food and water deprivation, rearing confined to one location of the cage, 5=marked cage tilt, change of cage mate, periodic loud noise andi stereotyped behav iour of licking and gnawing of cage wall, change in the size of the home cage, which change every confined to one location of the cage, and 6=assumption of few hours over a period of weeks or months. The effecti ve·· bizarre 'awkward disjunctive' postures. Classical ' and the ness of this procedure is usually monitored by tracking, newer anti-depressants, like mianserin, trazadone and over repeated tests, a decrease in the consumption of and/or fluoxetine, abolish isolation-induced behaviours. preference for a palatable weak (1-2%) sucrose solution. However, other end points, including parameters used in (D) Tail suspension test-This test is a variant of the the chronic unpredictable mild footshock technique, can behavioural despair test in which immobility is induced by also be investigated. Typical and atypical antidepressants simply suspending a rat or mouse by the taiI 2!. Unlike be­ can restore normal behaviour and perturbed physiological havioural despair, there is no hypothermia and the behav­ functions. The basic defect of this test, despite its utility in ioural changes last longer than the test period!. Mice pro­ evaluating antidepressant activity, and the reason for its vide better results and in a typical experiment a mouse is limited use, is the chronicity of the model and the difficulty hung on a wire in an upside down posture so that its nostril to set it up in a new laboratory. Considerable inbetween just touches the water surface in a container. After initial laboratory variations also preclude wide acceptance of the vigorous movements, the · mouse assumes an imrnobile model. Furthermore, the predictive, validity of this method posture and the period of immobility during a 5 min obser­ does not appear to be superior to the widely used Porsolt's vation period is noted. This test is a reliable and rapid or the learned helplessness tests. Some investigations have screening method for antidepressants, including those in­ questioned the reliability of this model since clinically ef­ volving the serotonergic system. However, MAO inhibitors 2 fective antidepressants may fail to reverse the chronic are usually inactive !. 27 stress induced physiological perturbations • (E) Chronic unpredictable stress---One of the major consequences of chronic stress is behavioural depression in (F) Separation models-The rodent model28 was 22 rodents . Several models have been used, mainly in rats, evolved on a primate model involving separation of the where, apart from chronicity of the stress, its unpredict­ mother from its progeny. The initial st!lge of 'protest' ability and inability to cope with the stressor, are major (agitation, insomnia, distress calls and screaming) is fol·· 22 23 factors . In one of the techniques , rats are exposed ran- lowed after 1 to 2 days by 'despair' (decreased normal ac- BHA ITACHARY A et al. : EV ALUA TION OF PSYCHOTROPIC AGENTS 121 tivity, loss of appetite, reduced social interaction and vo­ laboratories. Similarly, the specificity of tricyclic antide­ calization). TCAs selectively reduce the signs of despair. pressants, in terms of inhibiting noradrenaline or 5-HT re­ The same protocol is followed in rats and the model is re­ uptake, can be assessed by estimation of catechol-O­ garded as one of the best methods with face and construct methyltransferase metabolites, the levels of which will pre­ 34 validity28. dominate over the normal MAO metabolites • The role of

(G) Incentive disengagemenl--Rats are trained in a the serotonergic system, in general, and the 5-HT1A recep­ runway for food reward and then switched to non-reward tors in particular, in-the genesis of endogenous depression, situation (extinction of learning). Non-rewarded trials are appears to be a major factor. Most anti-depressants appear followed by augmented locomotor activity in the first to share the ability to increase 5-HT neurotransmission and 3s week, and by reduced locomotor activity in the second 5-HT1A receptor activity, by diverse mechanisms . There is 9 week. Although the method has good predictive validiti , compelling clinical evidence and experimental data in sup­ it has not proved popular because of procedural problems. port of this hypothesis which indicates that depression may (B) Disturbed circadian rhythm-Rodents are noctur­ be a consequence of malfunctioning of the median raphe nal animals, with high locomotor activity at night and rela­ serotonergic system I. The antidepressants which influence tively low activity during the day. Readjustment to a nor­ the serotonergic system include the reversible MAO A in­ mal circadian cycle of locomotor activity following rever­ hibitors (, brofaromine, cimoxatone, be­ sal of the light-dark cycle is expedited by subchronic (3-5 floxatone, , etc.), selective reuptake 9 days) administration ofTCAs • inhibitors (fluoxetine, fluxoamine, , ,

(I) Operant-reward test-This model is based on the , etc.), 5-HT1A ligands (, , premise that hungry rats, trained to press a lever for food , , etc.), 5-HT2 receptor (ri­ reward, find it difficult to desist from lever pressing if the tans erin, mianserin, ) and TCAs with predomi­ reward is dependent upon the waiting. In a typical testa, nant effect on 5-HT reuptake (, clortriptyline, animals are trained to wait for 72 sec between lever presses , , miinacipran, etc.). Mirtazepine, to receive food reward. TCAs characteristically increase which functions mainly as a central adrenergic alpha-2 the number of food rewards that the rat earns under this and increases noradrenaline release, is schedule since lever pressing is closer to the original wait known to increase the release of 5-HT as well. These re­ (72 sec) programme. The newer antidepressants and MAO ceptors are postulated to function as autoreceptors inhibit­ l inhibitors are also active in this testa. ing 5-HT release from its neurones . Likewise, some of the (J) Other methods--Some other techniques have been so-called 'selective' noradrenaline anti­ used to evaluate antidepressant activity which may not be depressants, like , , and possible to replicate in most laboratories, given the sophis­ , may not be entirely free from effects on the l ticated nature of the methods. They are : serotonergic system, including the 5-HT1A receptors . a. Olfactory bulbectomy--Bilateral destruction of the ol­ However, some of the recently developed antidepressants factory bulbs in rats, leads to impairment of learning, irri­ do not appear to influence the monoaminergic reuptake tability, hyperactivity and increased plasma corticosterone system. Thus, rolipram appears to stimulate noradrenergic 3 1 levels • The lag period for the appearance of the behav­ neurotransmission by, presynaptically, increasing nora­ ioural changes is 2-5 weeks and the technique has excellent drenaline synthesis and release, and, postsynaptically, by face validity, being attenuated by all classes of antidepres­ increasing cyclic-AMP concentration by inhibiting the en­ l sants. zyme phosphodiesterase . Similarly, acetyl-L-camitine, which has a structural similarity to acetylcholine and b. Intracranial self-stimulation-Experimental induc­ stimulates muscarinic receptors, has been found to exhibit tion of depression invariably leads to performance deficits antidepressant effect in elderly depressed patients I . in the performance of rewarded behaviours. Attempts were made to identify the brain centre responsible for reward, The major problem in evaluating an antidepressant in using implanted intracranial self-stimulation electrodes. the laboratory is choosing the appropriate tests in which the The ability of different classes of drugs to increase the fre­ new drug under investigation has to pass in order to be quency and duration of self-stimulation, provides an index clinically evaluated at a later stage. All the tests mentioned of antidepressant activity. Although, all classes of antide­ above have been shown to have reliable predictive and face pressants are effective, the model lacks specificityl2. validity inasmuch that they are capable of identifying a variety of antidepressants acting via diverse mechanisms. Conclusion However, each one of these tests have also been subjected The importance of pharmacological techniques in pre­ to criticism because of the 'false positives' and 'false nega­ clinical screening of antidepressants cannot be minimized. tives' encountered by different laboratories. It has been However, wherever possible, they have to be compli­ shown that anticholinergic and antihistaminergic drugs mented by corroborative biochemical paradigms. Thus, exhibited antidepressant activity in a variety of animal MAO inhibitor activity, including delineation of selective models of depression, whereas opiates and neuroleptics MAO A and MAO B inhibiting activity, using relevant appeared to accentuate the parameters used to assess de­ 3 substrates and enzyme sources ) , can be performed in most pression. Ipsapirone, a clinically effective antidepressant, is 122 INDIAN J EXP BIOL FEBRUARY 1999 reported to be inactive in several animal models of depres­ lessness and chronic stress tests. The acceptability of these, sion, including the chronic stress paradigm. Three: of the and the oilier tests used to evaluate antidepressants, is most popular rm.'fhods used to screen antidepressant activ­ based on a reasonable pharmacological profile, with rda­ ity have been subjected to extensive scrutiny and the results tively few false positive and false negative38 . In the absence are a mass of confusion. Whereas, the Porsoll's, learned of a universally acceptable animal model, it is advisable to helplessness and chronic mild unpredictable stres.s tests, use a battery of tests in screening for putative antidepres­ have been effective in delineating newer antidepressants, sant activi~9 . there are reports that they are of relatively low validity, The choice of methods in preclinical screening of puta­ particularly when used to screen atypical antidepressants36. tive antidepressants is a difficult one, given the variety of Depression has a number of temporal features, including methods available and the various ways in which the in­ spontaneous' recovery, the periodicity of recurrent unipolar vestigative drugs may function as an antidepressant. In our depression, and the alternation of depressive and manic own experience, with the limitation of resources, the fmt phases of bipolar depression. These aspects remain rela­ preclinical indication of antidepressant activity should tively unresearched in animal models. However, although come from reversal of the reserpine syndrome. When a more documentation is required to assess the validity and drug passes this initial test, the subsequent methods which reliability of animal models of depression, both in terms of are employed are the behavioural despair and learned the behavioural and neurochemical mechanisms involved, helplessness tests. Later, potentiation of 5-HTP, I-DOPA, there is no doubt that the Porsoll's test, and later the amphetamine and yohimbine actions are used to corrobo­ learned helplessness test, were breakthrough model:; which rate the earlier results and to get an indication of the neu­ supplanted the earlier used reserpine reversal test. These rotransmitter likely to be invoJved. Finally, the chroojc tests were simple and predictive, although lacking construct unpredictable footshock method is used to provide conclu­ validity. The introduction of the chronic mild stress models sive data39. The fmal proof will, however, come from clini­ was an improvement over these tests. However, despite the cal testing using well established parameters (HAM-D, observation that this test may have predictive validity, its MADRS, CGI severity). utility in antidepressant screening is limited by the chroni­ cally and, consequently, the time factor involved in the References 37 1 Sambunaris A, Hesselink 1 K, Pinder R, Panagides 1 & Stahl screening . While discussing the predictive validity and S M, J Clin Psychiat, 58 (SuppI6) (1977) 40. robustness Cjiterion of animal models of depression, it has 2 Bombardelli E & Morazzoni P, Fitoterapia, 56 (1995) 43. been commented that none of the existing models in use 3 Bhatttacharya S K & Satyan K S, Indian J Exp Bioi, 35 fulfil the expectations of the neurobiologist, in terms of (1997) 565. predictive, construct or face validi~8 . The answer to this 4 Willner P, TIPS, 12 (1991) 131. vexing problem is to search for pharmacological properties 5 Willner P, Psychopharmacology, 83 (1984) I. shared by antidepressants belonging to different chemical 6 Thiebot M H, Martin P & Puech A J, Br J Psychiat, 160 classes and acting via diverse neurochemical mechanisms. (Suppl 15)( 1992) 44. The utility of such common denominators can serve as an 7 Pertwee R G, Br J Pharmacol, 46 (1972) 753. .8 Mukhopadhyay M, Upadhyay S N & Bhanacharya S K, acceptable approach in evaluating antidepressant activity. It Indian J f-xp Bioi, 25 (1987) 761 . has been suggested that investigators should be more mod­ 9 Delina-Stula A, in Psychotropic agents, Pari I : An/ips)'­ est in advocating the utility of existing animal models of chotics and antidepressants, edited by F Hoffmeister & G depression and realise their shortcomings. In addition to Still (Springer-Verlag, New York) 1980, 505. their role in discovery of newer and more clinically effec­ 10 Ganguly S, Hota D, Goel R K, Acharya S B & Bhattacharya, tive antidepressants, animal models of depression a.re use­ S K, Indian J Exp Bioi, 34 (1996) 408. ful for providing insights into the neurobiology and patho­ 11 Puech A 1, Chemat, R, Poncelet M, Doare L & Simon P, physiology of depression. However, the data derived from Psychopharmacology, 75 ( 1981 ) 84. animal models are likely to be of value only to the extent 12 Tedeschi D H, Tedeschi E E & Fellows E 1, J Pharmacal that the models are valid, including consideration of pre­ Exp Ther, 126 (1959) 223. 13 Come S 1, Pickering R W & Warner B T, Br J Pharmacal, dictive validity (which concerns primarily the correspon­ 20 (1963) 106. dence between drug actions in the model and in the clinic), 14 Bhattacharya S K & Sen A P, J Neural Trans, 84(1991 J 24 1. face validity (phenomenological similarities between the 15 HorovitzZP,LifeSci,4(1965) 1909. model and the clinical disorder) and construct validity (a 16 Porsolt R D, Bertin A & lalfre M, Arch Pharmacodyn Th er, sound theoretical rationale for using the model). The basic 229 (1977) 327. desirable features in an animal model are that the model 17 Porsolt R D, Bertin A & lalfre M, Eur J Pharmacal, S 1 should respond appropriately to standard antidepressant (1978) 291. drugs, should employ physiologically realistic depression 18 Seligman H E P & Beagley G, J Comp Physiol Psychol, 88 simulating conditions and should model a core symptom of ( 1975) 534. 19 Martin P, Soubrie P & Simon P, Prog Neuropsychopharma­ the clinical disorder. Unfortunately, none of the existing col Bioi Psychiat, 11 (1987) I. models perform well against all three sets of validating 20 Einm D F, Morgan M 1 &Sahakian D, J Dev Psychobiol, 8 criteria, including the widely used Porsoll's, learned help- (1975) 553. BHA ITACHARY A et al. : EV ALUA TION OF PSYCHOTROPIC AGENTS 123

21 Chemat R, Thierry B, Mico J A, Steru L & Simon P, J 30 O'Donnell J M & Seiden L S, J Pharmacol Exp Th er, 224 Pharmacol, 17 (1986) 348. (1983) 80. 22 Stanford S C, Pharmacol Biochem Behav, 54 (1996) 211. 31 Klinger E, Psychol Rf?v, 82 (1975) I. 23 Katz R J, Pharmacol Biochem Behav, 16 (1982) 965. 32 Kokkinidis L, Zacharko R M & Predy P A, Pharmacol Bio- 24 Willner P, Towell T, Sampson D, Muscat R & Sophokleous chem Behav, 13 (1980) 379. S, Psychopharmacology, 93 (1987) 358. 33 Bhattacharya S K, Chakrabarti A, Sandler M & Glover V, 25 Armario A, Garcia-Marquez C, Adell A & Gelpi E, in Neurosci Lett, 199 (1995) 103 . Stress: Neurochemical and humoral mechanisms, edited by 34 Sulser F & Mobley P L, in Psychotropic agents, Part 1, G R Van Loan, Kvetnansky R, McCarty R & Axelrod J edited by F Hoffmeister & F Stille (Springer-Verlag, New (Gordon & Breach; New York) 1989, III. York) 1980, 471. 35 Porsolt, R D, Psychopharmacology, 134 (1997) 363. 26 Willner P, Psychopharmacology, 134 (1997) 371. 36 Willner P, Pharmacol Ther, 45 (1990) 425 . 27 Borsini F, Psychopharmacology, 134 (1997) 339. 37 Anisman H Q, Meral Z, Psychopharmacology, 134 (1997) 28 Everett C, in Proceedings of the first international sympo- 330. edited by Garratini & H sium on antidepressant drugs, S 38 Willner P, Papp M Z, in Antidepressants new pharmacologi- Dukes (Excerpta Medica, Amsterdam) 1966, 164. cal strategics, edited by P Skolnick (H umana Press, N.J .) 29 Steru L, Chemat R, Thierry B, Mico J A, Steru M, Simon P 1997,213 . & Porsolt R D, Prog Neuropsychopharmacol Bioi Psychiat, 39 Bhattacharya S K, Chakrabarti A, Chatterjee SS, Pharma- II (1987) 659. copsychiat, 31 (Supp I) (1998) 22.